Designing the Virtual Statehouse
- By Elizabeth Sikorovsky
- May 31, 1998
Have you ever wondered why state capitols have domes? Or why some bank entrances are flanked by tall columns? Over time, architects have crafted elaborate codes for designing city structures and spaces. These codes govern the way 3-D space is carved out to create a sense of grandeur or to direct the movement of people through buildings.
But architectural codes may have an equally important role to play in the development of electronic government, according to two faculty members at the Massachusetts Institute of Technology. In a new course called the Virtual Statehouse introduced this year, two teachers challenged students to integrate the disciplines of architecture, urban planning, public policy and computer science to improve government services online.
Four World Wide Web-based student projects emerged from the course: a virtual help desk, an online courtroom system, a geographic data search tool and a tour of the Massachusetts statehouse (see sidebars)."Architecture is a people-centered field rather than a technology field," said William Mitchell, dean of MIT's School of Architecture and Planning, who co-taught the course with Dan Greenwood, deputy general counsel for the Information Technology Division of the Massachusetts state government. "The key issue in our class was to explore how you can establish in a natural and graceful way what are the conventions and the consequences of your actions on a government site.
"In the physical world, if you see a building with a big gold dome on it, you know where you are," he added. Yet, "online environments have been extraordinarily crude in making clear the expectations and consequences of our actions online."
Mitchell is author of City of Bits, a book that describes a city configured in online services and networks instead of buildings and roads. In his book, Mitchell argues that cryptic commands and alienating interfaces can be as much a barrier to entry to government services as a flight of steps is to a person in a wheelchair. Similarly, just as a dome on a capitol building can convey a sense of importance and permanence, a Web site can confirm or undermine a user's impression of state government.
Helping people to feel "at home" online should be a top priority for government organizations, Mitchell said. Indeed, the successes and failures of important public spaces, such as New York's Central Park, depend on how visitors are welcomed, cued and directed as they navigate through an area. These issues hold true for the online world.
Civic Web design is vitally important, said Mitchell, because governments mark their approach to civil liberties in the way they build their public spaces. In his book, Mitchell warns of "electronic Jakartas" or "well-connected, well-serviced, fortified enclaves of privilege surrounded by miserable hyperghettos," where service is poor and content is limited. Free access to information, sponsored by government, may help to prevent that.
Although some of these lessons are basic, Mitchell said, it is easy to lose sight of simple design principles in the technology thickets of the Internet. "A huge amount of the work has focused on extraordinarily technical issues, but what has seemed to be missing has been an overall design framework," for state government online service, he said.
A general lack of design purpose has made navigating government Web sites like driving through an unincorporated area with poor zoning. Citizens "navigate in a wide, fractured environment where there's a huge array of data," Greenwood said. One of the drawbacks of this model is that citizens cannot always tell what the legal ramifications are of being on a particular site.
This problem is especially important in government because it supports so many activities, from publishing to contract signing, to discussions with government leaders. "How do you know when you're in a situation where you have to be quiet? When you can be loud?" Greenwood asked. "You can have airtight technology, but legally the questions are about making sure a person signing an electronic document knew what he signed. You need to bring in the designers for that." More consistent, clearer design online is crucial for government, he said.
Addressing design issues is important as state and local governments move into the "next phase" of Web development, said Theresa Pardo, project director for the Center for Technology and Government at the University of Albany in New York and an expert in public-sector IT. "Functionally, I think absolutely we've moved past the point of 'as a commissioner, I want to have a Web site.... I don't know why, but I want one,' " she said. "We have found that unless the Web sites were functionally driven, they languished."
At the same time, she believes agencies have been faster in improving how they share information with other agencies than in how they deliver information to the public. "We're finding that the value of sharing information on an interagency level is more advanced. We know our audience better. It's easy to talk about the Web and what it can do, but the hard part is figuring out a need that the customer has and finding a way to improve it via the Web."
Mitchell and Greenwood are teaching a follow-on course to the Virtual Statehouse project that covers commercial as well as government applications. Mitchell sees courses such as this as crucial to creating a valuable, democratic, easy-to-navigate Web. "We're all very good at understanding what's expected of us in architectural spaces," he said. "But we don't have that same understanding in virtual spaces."
-- Elizabeth Sikorovsky is a free-lance writer based in Boston.
For his project, Carleton Tsui, a graduate student in MIT's School of Urban Planning, proposed a virtual help desk that visitors to an online site can use if they get lost on the Web. The concept integrates Web and Internet Relay Chat (IRC) technology to accommodate the tradition of walking up to an information booth and asking questions.
Tsui's inspiration for the project came by imagining what citizens find when they walk into a large unfamiliar building. "When you wander into a government building like that, it's usually...massive," he said. "And if you can't find this or that office, more often than not you can go to an information center and get directions. The trick is to duplicate that in terms of a Web site."
Tsui's proposal calls for an actual human being to be at the virtual desk, ready to answer Internet visitors' questions with IRC. A user could ask directions on how to reach a particular site or service, and the help-desk attendant could answer the question in IRC while simultaneously producing the requested Web site on the user's screen.
His plan mimicked a real-world help desk even more closely by allowing information to pass among users "standing in line." Users waiting in line could watch a dialog between the help desk and the user being served. That way, users could have their questions answered before they had a turn to chat with the help desk. Users waiting in line could also talk to one another. The key to applying architectural models to cyberspace, Tsui said, is to identify specific interactions that work well in the physical world and then to adopt them in the virtual world. "Try to find the value-added first," in the physical setting, Tsui said. "What is it that you like about a particular interaction?"
Thinking this way might help avoid the problem of transferring physical models to the online world in ways that do not fit the medium. "There's a danger in transplanting physical designs straight to the Web," Tsui said. "Our first thought was to put an actual architectural model of the statehouse online. Based on student participation, I think that concept changed a lot.
"I don't think transplanting architecture to the Web in a literal sense is the way to go," he continued. "It uses up a lot of bandwidth, for one thing. You can have a library, a market, trees, garbage cans, but what is the value of doing that? More often than not, [the metaphor] starts to break down."
Tsui warned that in an age where search engines can parachute a user into a site without going through a main home page, creating an elaborate "front entrance" to a state Web site may not be worthwhile. "Search engines are an entry point to Web sites now," he said. "Chances are that users won't come in through the front door."
Another tactic in designing new online government services is thinking about how different technologies can be combined to add value. In his project, Tsui addressed one of the problems he sees with the current evolution of the Internet: the fact that basic services are not integrated. "Right now, we have newsgroups, e-mail, chat and the World Wide Web," Tsui said. "There are a lot of things in front of us all the time that, if you meld [them] together, create new solutions."
- Elizabeth Sikorovsky
Tapping Today's Systems
MIT urban studies graduate student Raj Singh stressed the importance of using the metaphor of the 3-D world carefully. Often, translating solutions that work in the physical world to the online world results in kludgy applications that slow down the user. "It seems to me that most people are trying to get quick access to information," Singh said. "They want to get in and get out as quickly as they can. With that in mind, I found it hard to take advantage of a 3-D concept on a Web site."
However, like Tsui, Singh suggested that information could be organized in ways similar to models in the physical world. "One metaphor that works well is the library system," he said. "When you do a search in a library, you may find some books, and then when you go to find those books, you see other related ones on the shelf. That works because you have skilled librarians figuring out what books should be shelved next to each other in the Dewey Decimal System. I haven't seen that type of organization online."
Singh suggested that using physical metaphors is extremely useful when the data relates to the physical world. "[On] a government Web site, I don't see a generic 3-D metaphor fitting everywhere, but there are certain applications geared toward physical planning that are perfect. If the nature of the information is geographic, I definitely see a connection."
Singh's project allows citizens to search for geographic and environmental data about their own communities. Citizens use an online map of the state or county to pinpoint the region they want to research. By pointing to a region, the system zeroes in on a particular online data set and offers it to the user. Singh is developing his project-putting more than 200 CDs of environmental and geographic data online-for Massachusetts.
"The Massachusetts project is unique in its sheer magnitude," said Singh, explaining that his team is taking an existing application that was developed more than three years ago for Environmental Systems Research Institute Inc.'s Arc-View and porting it to the Web. The database is in the tens of gigabytes, and the geographic range is the entire commonwealth of Massachusetts, Singh said. In addition, he is setting up a similar but smaller project for a city in Florida.
Singh has plans for even more elaborate applications to bring zoning and construction approval process online. "My grand idea is that a developer might submit his building plans electronically and then people could submit their 'what if' scenarios," he said. "Maybe they could draw over a copy of the plan online and submit comments. With that sort of thing, you're not just talking to the planners, you're talking to everyone."
- Elizabeth Sikorovsky